WorksPhysics and Astronomy
Analysis of -wave dynamics behavior in multi-body and decays
Abstract
The state has a sizable decay width, which have longstanding puzzle in treating as a final state in theoretical calculations may lead to systematic deviations in the predicted observables. To address this issue and further explore the internal structure of this resonance, we study the multi-body semileptonic decays and . Within the approximation where the resonance dominates, we study the decay in the QCD light-cone sum rules framework. The corresponding transition form factors (TFFs) are calculated by constructing a correlation function with a right-handed chiral current. As the main nonperturbative input for the TFFs calculation, the transverse twist-2 light-cone distribution amplitude (LCDA) of the resonance is constructed using the light-cone harmonic oscillator model, and the resulting LCDA exhibits a single-peak behavior. At the large-recoil region, we obtain the TFFs and , together with the corresponding TFFs ratios and . We then further study the complete decay by incorporating the resonant line shape and obtain the corresponding observables. The results show that the branching ratio of the decay is about of that obtained when is treated as a final state. We also account for the subsequent decay and further predicted the branching ratio of the decay . These results may provide useful references for future experimental measurements and further theoretical studies.
Introduction
In the low-energy region of strong interactions, information about the internal structure of excited hadronic states is often difficult to extract directly from their complex invariant-mass and angular distributions. This is particularly relevant for axial-vector states, which may have rich internal configurations and sizable finite-width effects. Identifying these effects through experimentally accessible observables is therefore important for understanding the structure of light hadrons. Semileptonic decays of heavy-flavor -mesons provide a useful platform for studying weak transitions and hadronic transition dynamics [1]. In such processes, the resonance of interest can be produced through a relatively clean short-distance weak transition, while its properties are subsequently encoded in the experimentally measurable multi-body final states through its strong decay.
At present, several axial-vector states are known to have sizable decay widths. Both experimental and theoretical studies of these states therefore often rely on the multi-body final states formed by their decay products. Semileptonic decays provide a relatively clean environment for such investigations. Since the final-state lepton does not participate in strong interactions, the hadronic dynamics is mainly governed by the weak-current-induced transition and the subsequent strong interactions among the final-state hadrons. Compared with purely hadronic decays, semileptonic processes generally suffer from reduced hadronic backgrounds and are therefore better suited for studying the spectroscopic properties and internal structure of axial-vector resonances [2, 3]. Experimentally, in 2024, the BESIII Collaboration reported the first observation of with a statistical significance of , and also found evidence for with a significance of [5]. In both decay channels, the resonance was reconstructed through the final state. These measurements provide direct experimental motivation for further studies of semileptonic decays involving the resonance. On the theoretical side, the theoretical side, the decays were systematically studied within SU flavor symmetry, where and , , , and denote axial-vector, tensor, vector, and pseudoscalar states, respectively [4]. It was shown that both axial-vector and tensor resonances can contribute to such decay modes, indicating that multi-body semileptonic decays provide a useful approach for studying resonance structures in hadronic final states. In addition, lattice QCD studies indicate a strong coupling of the resonance to the -wave channel [6], further supporting studies of this resonance through the final state. According to the PDG [1], the -meson decays predominantly through , with a branching fraction of . Moreover, its natural width of MeV, is much smaller than that of the resonance, MeV. Therefore, the resonance can be experimentally reconstructed through the corresponding four-pion final state, and the complete decay chain can be written as . Despite these experimental and theoretical developments, systematic studies of the decay remain limited. A detailed investigation of this process is therefore worthwhile for studying the transition dynamics, the finite-width effect of the resonance, and the resonance structure of the final state. Axial-vector states possess a rich internal structure. They have the same total angular momentum and parity, but different states show different properties because of differences in symmetry, composition, and dominant decay channel. For non-strange systems, and are classified as -states, whereas , and are classified as -states. The strange states and have no definite or -parity, they involve mixing between the and basis states. According to the latest PDG compilation [1], The mass of the state is MeV, its average width is MeV and its main decay channels are [7]. Which compared with other axial-vector states, The mass and width of are MeV and MeV, respectively. The dominant decay channel for both and is [8, 9, 10, 11], and both states have relatively large decay widths. The widths of ( MeV) and ( MeV) are relatively narrow, and they mainly decay into [12, 13, 14] and [12, 13, 15, 16, 17]. Although and have similar masses, they differ in their internal quantum numbers and -parities. The considerable width of resonance makes it an important tool for studying the multi-body dynamics of axial-vector resonances.
Currently, the internal structure of axial-vector states is still not fully understood. In particular, the internal structure of state has not yet been completely determined. The state was first observed by the HBC in 1963 [18]. Before 1994, studies on the resonance mainly focused on processes such as [19, 20], [21], and [22, 23, 24, 25, 26, 27, 28, 29, 30], with the main goal of measuring its mass and width. In the processes at rest, the (then called the ) was clearly observed. Its mass, width, and spin-parity were determined to be approximately 1200 MeV, 100–130 MeV, and , respectively [19, 20]. For the collision experiment in 1974, the decay channel was analyzed at 5 GeV/c, and the mass and width of the resonance were measured, yielding values similar to those obtained in [24, 26]. It is indicated by these results that the is a resonance with a significant width. In the traditional quark model, is often described as a quark-antiquark pair in the state [31, 32]. Lattice QCD calculations show that the resonance can be effectively produced by an operator composed of a quark and an antiquark. This indicates that it has a significant component [33]. The calculations also show that the interaction between and the system is also important [6]. Some theoretical studies have proposed different explanations. For example, in the diquark-antidiquark color-spin model, can be regarded as a candidate state of the tetraquark state [34]. Based on the above discussion, more theoretical support is given to as a state, and the subsequent study in this paper is carried out on this basis.
For the decay , the decay dynamics involves both the weak transition and the subsequent strong decay . These two parts are connected through the propagator. To describe the resonance line shape and the finite-width effect near the threshold, we adopt a Breit-Wigner form with a mass-dependent width [35]. The nonperturbative dynamics of the weak transition is encoded in the transition
| Vector state | symmetric | symmetric |
| Axial-vector state () | symmetric | antisymmetric |
| Axial-vector state () | antisymmetric | symmetric |
Table 1. The LCDAs of vector and axial-vector states exhibit symmetric or antisymmetric behavior under the exchange of quark and antiquark momentum fractions [44].
form factors (TFFs), which therefore serve as key hadronic inputs for calculating the differential decay rates and other physical observables. A reliable determination of these TFFs is essential for a quantitative description of the four-body decay. In particular, precise knowledge of the TFFs provides an important theoretical basis for studying various observables of this decay and for assessing the finite-width effect of the resonance.
Theoretically, many methods are available for studying heavy -meson to the light axial-vector TFFs, including perturbative QCD (pQCD) [36], QCD light-cone sum rules (LCSRs) [37, 38], and the covariant light-front quark model (CLFQM) [39, 40]. For pQCD approach in 2009, the factorization scheme which retains the transverse momentum is adopted where the TFFs were calculated TFFs at large-recoil region and whole physical region directly. Branching ratios for -meson decays to different axial-vector states is also given, most of them are at the level. It was also pointed out that the sizes of the TFFs and different helicity components are very sensitive to the distribution amplitudes of the axial-vector state [36]. Implementation of the CLFQM method, a scheme was adopted in which the calculations were carried out in the spacelike region and then extended to the timelike region through a parametrization. Within this scheme, the TFFs were treated separately for the spacelike and timelike regions. It was also pointed out that the internal structure of axial-vector states could significantly affect the related TFFs predictions [40]. For the LCSR method in 2008, the vector and axial-vector TFFs were calculated for decays to -wave axial-vector states, with order corrections included. The differential decay widths and branching ratios for the subsequent were measured [37]. A basis for evaluating the reliability of TFFs predictions can be provided by comparing different results [41]. In this work, we use the LCSR method, it is suitable for the low to mid -region required for this process, the LCDA of resonance can be used to describe the relevant nonperturbative effects. The uncertainties of nonperturbative input can also be evaluated. This method was developed from 1980s. It can be regarded as an extension of the SVZ sum rules to the light cone [42, 43].
To accurately determine the behavior and precision of the TFFs in the transition, the leading-twist LCDAs of the state play a key role, which contain information about the momentum distribution of quarks inside the hadron, Therefore, we also pay close attention to the behavior of its twist-2 LCDA amplitude. Within the LCSR framework, for axial-vector state, the LCDAs include both longitudinal and transverse polarization states, which are described by chiral-even and chiral-odd operators, respectively. Regarding LCDAs, opposite symmetry properties are exhibited by the chiral-even and chiral-odd cases under quark-antiquark momentum exchange [44]. From the QCDSR [44] sum rules, it can be known that the chiral-even two-particle LCDAs of axial-vector state are antisymmetric, while the opposite holds for the chiral-odd LCDAs. The symmetric and antisymmetric properties of the twist-2 LCDAs of axial-vector states under the exchange of the quark momentum fractions are listed in Table I. For the longitudinally polarized one couples through the axial-vector current , while the transversely polarized state is described by the tensor current . We use the right-handed chiral current to construct the correlation function and obtain the TFFs. We choose the chiral current over the conventional current to construct the correlation function. Within the present QCD light-cone sum rule framework, this choice helps suppress unwanted resonance contributions, allows some strong interaction corrections to cancel, and reduces the sensitivity of the results to higher-twist distribution amplitudes [45, 46, 47, 48], These features can improve the stability of the calculation and the precision of theoretical predictions [47, 48]. For the transverse leading-twist distribution amplitude , which gives the dominant contribution, the light-cone harmonic oscillator (LCHO) model is adopted to construct it. This model is based on the Brodsky-Huang-Lepage (BHL) prescription, which connects the equal-time wave function of the meson in its rest frame to the light-cone wave function in the infinite momentum frame [49, 50, 51], it provides an intuitive and effective framework for modeling the internal structure of mesons. This model has been successfully applied to pseudoscalar mesons , , [52, 53, 54], scalar states , [55, 56, 57] and vector states , , [58, 59, 60]. At the same time, for axial-vector states, the LCHO model has also been successfully used to construct the longitudinal twist-2 LCDA of . It has also been applied to study the TFFs and semileptonic decays [61]. This shows that the LCHO model can provide an effective parametrization for the nonperturbative structure of axial-vector states. Based on this, in this work, we adopt the LCHO model to construct the transverse twist-2 LCDA of as an input parameter. Then, we use LCSR to study the semileptonic decay process with a four-body final state.
The structure of this paper is organized as follows: In Section II, the decay width of the decay and the theoretical framework for the calculation of the TFFs by means of the LCSR method are first briefly introduced. This includes the detailed procedure for the construction of in the LCHO model, the method by which the correlation functions are built with the right-handed chiral current, and the expressions for the TFFs. Section III presents the numerical results and analysis, covering the behavior of the LCDAs, the shapes of the TFFs, the differential decay width, the branching fraction, Finally, the paper concludes with a summary and outlook.
Theoretical Framework
A general expression for the differential decay width of decays has been derived in Ref. [62]. It is formulated in terms of the five-variable Cabibbo-Maksymowicz parametrization [63] and has been widely used in theoretical studies of processes. In this work, we adopt this general formalism to describe the decay . The effective weak interaction amplitude for the transition can be written as
where is the Fermi coupling constant. For a positively charged final-state lepton, the leptonic current is defined as , which can be calculated perturbatively. The hadronic current describes the charged-current matrix element of the transition. Its nonperturbative dynamics is parametrized in terms of a finite set of independent TFFs through a Lorentz-covariant decomposition of the matrix element. For the decay, the weak production of the resonance is followed by the strong decay . These two parts are connected through the propagator. The corresponding resonance line shape is described by a Breit-Wigner form with a mass-dependent width, which allows the finite-width effect of the resonance to be incorporated. The differential decay width is given by , where denotes the four-body Lorentz-invariant phase space. For the process , the phase space can be factorized into the successive two-body phase spaces , , . After integrating over the overall orientation of the spinless initial meson and neglecting the charged-lepton mass, the final-state kinematics can be completely described by five variables, as shown in Figure 1. The five variables are the invariant mass squared of the system, ; the invariant mass squared of the system, ; the polar angles and defined in the and rest frames, respectively; and the azimuthal angle between the two decay planes. Here, , , and denote the four-momenta of the meson, the pion, the charged lepton, and the neutrino, respectively. In terms of these five variables, the differential decay width can be written as

Figure 1. Definition of angular variable, among them, the three-momenta of in the rest frame of the virtual and of rest frame are denoted by and , respectively. The unit vector along the direction in the rest frame is set as , and the unit vectors along the components of and that are perpendicular to are defined as and , respectively.
where and . Here, denotes the magnitude of the three-momentum of the system in the -meson rest frame, while denotes the magnitude of the three-momentum of the (or equivalently the ) in the rest frame. The angular function can be decomposed into nine angular terms, –, which depend on , , and [62]. The physical phase-space region for the decay is determined by energy-momentum conservation. The allowed kinematic ranges are and . For a fixed invariant mass squared of the system, the maximum value of the dilepton invariant mass squared is therefore determined by . To obtain the double differential decay width with respect to the hadronic invariant mass squared and the dilepton invariant mass squared , the angular variables appearing in Eq. (2) need to be integrated over. Upon integration over the full angular phase space, only terms with nonvanishing angular integrals survive, while the remaining interference terms vanish. Thus, the double differential decay width is obtained as:
Here, , , and denote the corresponding P-wave amplitudes. In the quark model, the resonance is classified as a state, corresponding to a P-wave axial-vector state [1, 31]. Therefore, only the P-wave contribution is retained. According to previous studies, P-wave amplitudes entering the angular distribution can be directly related to the helicity amplitudes describing the weak transition [62, 64]. Therefore, the partial-wave amplitudes () can be expressed in terms of the helicity amplitudes as follows:
Here, . We take the total decay width GeV and the branching fraction [5]. The quantity denotes the magnitude of the three-momentum of the meson in the resonance rest frame, evaluated at the resonance pole mass. In addition, the line shape of the resonance can be described by a Breit-Wigner parametrization with a mass-dependent width [35]. The corresponding line-shape amplitude is given by [35, 65]:
The denote the mass dependence width have the following form
The is the Blatt-Weisskopf damping factor. The three helicity amplitudes can then be related to the axial-vector and vector TFFs and ,
where the TFFs are evaluated at the pole mass of the resonance and have no explicit dependence on . We find that contributes only to , whereas contributes only to . By contrast, contributes to all three helicity amplitudes, and . At large values of , the contribution from becomes increasingly important.
Based on the above theoretical analysis, the three TFFs serve as the main nonperturbative inputs for the calculation. A precise understanding of the behavior of these TFFs is essential for two reasons: it enables accurate predictions of the decay branching fraction and provides a theoretical basis for extracting the CKM matrix element from experimental measurements. The TFFs are indispensable for calculating the above observable physical quantities. Under the LCSR framework, the TFFs are derived from the vacuum-to-meson correlation function, and their dominant form is given as:
where and . This choice removes contributions from chiral-odd LCDAs, so that the non-perturbative inputs in the TFFs are mainly determined by the chiral-even LCDAs. Following the standard QCDSR, we insert a complete set of states with the same quantum numbers as the meson into the correlation function defined in Eq. (8). In this step, we use the matrix element , where is the -meson decay constant, together with the Lorentz-invariant TFFs. After isolating the pole term of the lowest pseudoscalar -meson and using a dispersion relation to model the contributions from higher resonances and the continuum, we obtain the hadronic representation of the correlation function. In the spacelike -region, the correlation function defined in Eq. (8) can be calculated using QCD. Specifically, we perform an operator product expansion (OPE) near the light cone, where . This expansion is valid when and the momentum transfer satisfies . We then apply a Borel transformation with respect to to enhance the ground-state contribution relative to higher resonances. The continuum contribution is treated using quark-hadron duality, allowing the OPE representation to be matched to the hadronic representation. Within the LCSR framework, this procedure gives the following general expressions for the TFFs:
where corresponding to , and in the axial-vector and vector TFFs. For , in the axial-vector TFFs satisfies , where is an auxiliary TFF. The explicit expressions for all parameters appearing in the analytical formulas for the TFFs are presented in Appendix Appendix A.
For the two-particle twist-3 LCDAs and , the analysis is more complicated than that for pseudoscalar states. This is because axial-vector states have different polarization states and their mass cannot be ignored. As a result, handling these functions becomes more difficult. In 2007, the twist-3 LCDAs of axial-vector states were systematically studied [44]. Using a conformal partial-wave expansion and the QCD equations of motion, which showed that and are not independent. Instead, they can be expressed in terms of simpler quantities: the Wandzura-Wilczek (WW) part derived from the twist-2 LCDA, the three-particle LCDAs, and the terms from quark mass corrections. Among them, the contribution from the three-particle LCDAs is very small, we can safely ignore it. The quark mass correction terms involve a factor , which depends on the quark mass inside the meson. For the resonance, this factor is basically zero [66]. As a result, the WW approximation is adopted in this paper. Under this approximation, the two-particle twist-3 LCDAs and can be approximately expressed as convolution integrals of the transverse twist-2 LCDA . The explicit expressions are given in Ref. [44]. Therefore, the twist-3 two-particle LCDAs required in the subsequent calculations can all be determined from under the WW approximation. Using the higher-twist LCDAs given in Ref. [67] within the simplest self-consistent approximation satisfying the QCD equations of motion, we obtain the expressions for , and . The values of the coefficients entering these expressions are given in Ref. [68].
The correlation functions in Eq. (9), together with the WW approximation, indicate that the dominant contributions to the TFFs arise from the transverse twist-2 LCDA of the resonance. We therefore employ the light-cone harmonic oscillator (LCHO) model to construct this LCDA. The twist-2 LCDA can be obtained from the corresponding light-cone wave function (LCWF) by integrating over the transverse momentum. The relation between the LCDA and LCWF is given by:
where and represents the transverse momentum of the resonance. Following the Brodsky-Huang-Lepage (BHL) description [49, 50, 51], the light meson wave function can be broken down into a spin wave function and a spatial wave function , which can be written as
In this expression, the correct relativistic spin structure is provided by , and the longitudinal and transverse momentum distributions of the quark-antiquark pair are described by . The symbols and refer to the helicities of the quark and antiquark, respectively. The function describes the coupling structure of the quark and antiquark spins, and it is determined by the Wigner Melosh rotation [51, 69, 70]. The spin-orbit coupling information is incorporated into the corresponding projection factors and wave-function definitions through the Wigner-Melosh rotation. For convenience, then, takes the simplified form , where is the constituent quark mass. Brodsky-Huang-Lepage (BHL) established a scheme by comparing the off-shell propagators in the rest frame and the light-front representation. In this scheme, the harmonic oscillator wave function in the rest frame is mapped to the light-front momentum space [71]. Therefore, the spatial wave function can be split into two parts. The first depends only on the longitudinal momentum fraction and is written as . The second depends on and comes from the harmonic oscillator solution that describes the meson in its rest frame with , where is the normalization constant, is the harmonic parameter. When quark-spin effects are taken into account, the Lorentz boost induces an additional longitudinal correction to the LCDA, which is incorporated through the function . This function can be expanded in terms of Gegenbauer polynomials as with , where is a Gegenbauer polynomial and plays an important role in determining the longitudinal part of the distribution. Finally, by integrating over the transverse momentum using Eq. (10), we can determine the twist-2 LCDA within the LCHO model. result is
where denotes the error function, it is obtained by integrating over the transverse momentum and rewriting the integral in Gaussian form. The integral result can be expressed as the difference between two error functions, with the first evaluated at the upper limit of the transverse-momentum integral , and the second at the lower limit . The difference between these two error functions represents the contribution to the twist-2 LCDA from transverse momentum below the factorization scale . This form in Eq. (12) has two useful features. First, it approaches the asymptotic form in the limit of large . Second, it is exponentially suppressed near the endpoints, thereby suppressing endpoint singularities in convolution integrals. In addition, three model parameters remain to be determined , , and . To do that, we use some extra conditions: (i) The normalization condition of the twist-2 LCDA for the resonance; (ii) The average value of the transverse momentum squared GeV [72]; (iii) The first Gegenbauer moments of the twist-2 LCDA. For this purpose, we adopt the value at the initial scale GeV [44]. The above theoretical framework lays the foundation for calculating the physical observables considered in this work.
Numerical Analysis
Before performing the numerical analysis, we adopt the following input parameters from PDG[1]. The masses of charged and neutral -mesons and -resonance are taken as MeV, MeV and MeV, respectively. For the quark masses, we use GeV, MeV and MeV. The masses of the and mesons are taken as MeV and MeV. The decay constant GeV [48], and GeV [44]. For transition, we take the typical process scale GeV. The transverse twist-2 LCDA plays a key role in determining the TFFs, both directly and indirectly. We therefore investigate the behavior of the transverse twist-2 LCDA firstly. Using the three conditions mentioned before: normalization, average transverse momentum squared, and Gegenbauer moment, we can fix the unknown parameters for . To obtain the LCDA parameters at any scale, the Gegenbauer moment can be evolved from initial scale to typical process scale by
| +25.877 | +0.182 | −0.574 |
| +29.671 | +0.012 | −0.626 |
| +32.401 | −0.175 | −0.662 |
| +24.703 | +0.152 | −0.584 |
| +27.604 | +0.022 | −0.623 |
| +29.910 | −0.118 | +0.653 |
Table 2. The three model parameters of resonance transverse twist-2 LCDA, i.e. , , and at initial and typical scale . From which the upper limits, central values, lower limits are present, respectively.
using the renormalization group equation
The one-loop anomalous dimensions are given by [73, 74]:
Here, , where denotes the number of active quark flavors. The resulting numerical values are listed in Table II. The resulting behavior of is shown in Figure 2, where the uncertainty comes from the error in . The transverse twist-2 LCDA obtained in this work is presented in Figure 2, with the solid curve representing the central value and the shaded band indicating the corresponding uncertainty. For comparison, the QCDSR prediction based on the truncated Gegenbauer expansion is also included [44]. Our result is in good agreement with the QCDSR [44] prediction, with both exhibiting a single-peak structure symmetric about , and agreement with the theoretical expectation. Our numerical result is slightly higher than the QCDSR [44] prediction, but it is still within the error range. This discrepancy may arise from differences in model construction, the choice of input parameters, and uncertainty propagation. Overall, the results presented in Figure 2 support the applicability of the LCHO model in constructing the transverse twist-2 LCDA of the resonance and provide a useful theoretical input for the subsequent calculation of the TFFs. In addition, the calculation of the TFFs for the transition requires two important input parameters: the continuum threshold and the Borel parameter . In the QCD sum rule approach [76, 77, 78], we use the following criteria to determine them:

Figure 2. The behavior of in LCHO model within uncertainties. The dashed line represents the previous QCDSR [44] prediction.
The contribution from the continuum is less than 35%.
The contribution from higher-twist LCDAs does not exceed 5%.
The value of the TFFs remains stable within the Borel window.
Based on the constraints mentioned above, we choose the continuum thresholds as and . The corresponding Borel parameters are chosen as , , , and . Using the above parameters and Eq. (9), we can determine the TFFs for at large recoil. The results are presented in Table III and compared with the theoretical predictions from pQCD [36], LCSR(I) [75], and LCSR(II) [38]. The uncertainties in our predicted TFFs are controlled within the range of 10% − 15%. Our results for agrees well with the pQCD prediction [36], while our result for is close to the LCSR(I) prediction [75]. For the remaining TFFs, our predictions have some differences compared to LCSR(I) [75] and LCSR(II) [38]. These discrepancies may arise from differences in the choices of chiral currents and the input distribution amplitudes adopted in the respective approaches. Additionally, we can define two particularly interesting ratios in the large-recoil region, and . These ratios can be obtained without relying on the total decay width or the CKM matrix elements, and they effectively remove common uncertainties from the overall normalization of the TFFs. This makes them more sensitive to differences in how various models describe the internal dynamics of hadrons. These ratios are listed in Table IV, together with other theoretical predictions. Different theoretical approaches employ different methods and input parameters. In this work, we obtain the central values and
Table III. The TFFs , and at the large recoil . The predictions of other theories are used for comparison.
Table IV. The ratio and of TFFs. For comparison, other theoretical results are also included.
. In particular, our prediction for differs appreciably from those of pQCD [36], LCSR (II) [38], and LFQM [40]. In general, these approaches predict smaller values of than our result. The predictions for also differ to varying degrees among different theoretical approaches, indicating that the relative magnitudes of the TFFs in the transition remain somewhat model-dependent. To facilitate the comparison of theoretical predictions for and , we present the corresponding confidence ellipse in Figure 3, which illustrates the joint probability distribution of

Figure 3. The confidence ellipses for the TFFs ratios and in the decay are shown. The empirical value is adopted from Ref. [35] to determine the tilt of the ellipses. We also compare our results with predictions from other theoretical approaches.
the two ratios obtained in this work. The comparison reveals differences among theoretical predictions for the three TFFs at the large recoil point. To assess the reliability of these results, one can examine the overall behavior of the TFFs across the full kinematic range.
The LCSR method used to calculate the TFFs is reliable mainly in the low and intermediate regions, but the decay width and branching fraction require knowledge of how the TFFs behave across the entire physical . To analytically continue the TFFs to the full allowed physical region, we employ the simplified series expansion (SSE) [79], given by
Where is a coefficients, the conformal variable and the reference point follow the standard definitions given in Ref. [79]. This parametrization provides a systematic and largely model-independent description of the semileptonic TFFs. It allows the near-threshold behavior of the form factors to constrain the expansion coefficients while preserving their analytic structure.
Additionally, the real coefficients are determined by requiring the quality of extrapolation to be less than 1%. Using the input values reported in Ref. [80] and by the PDG [1], we obtain the fit parameters and the fit quality for each form factor, as listed in Table V. In this case, all LCSR parameters are set to their central values. From the table, the values of corresponding to the four TFFs are found to be 0.030%, 0.806%, 0.007% and 0.145%, respectively, all of which are less than 1%. Among them, has the best fit quality. Although the value for is relatively large, it is still significantly lower than 1%. This indicates that adopted parametrization accurately reproduces the LCSR results for the TFFs within the fitting interval, providing a reliable basis for further investigating their behavior over the full kinematic region. Using the parametrization described above, we obtain the dependence of the TFFs over the full physical kinematic range, as shown in Figure 4. In Figure 4, the solid curves represent the central values, while the shaded bands indicate the corresponding uncertainties. We also compare our results with the predictions from LCSR (I) [75], LCSR (II) [38], and pQCD [36]. The different theoretical approaches exhibit noticeable differences in the extrapolation behavior of the TFFs. Our extrapolated TFFs show varying degrees of agreement with predictions from different theoretical approaches. In the low and intermediate regions, our result for agrees well with the pQCD prediction [36]. Similarly, our result for is close to the LCSR (I) prediction [75] in these regions, although the discrepancy gradually increases with . In our predictions, both and remain positive and increase monotonically with . This behavior differs markedly from the predictions of LCSR (I) [75] and LCSR (II) [38]. In contrast, the pQCD predictions [36] exhibit a similar dependence to that obtained in this work, although their overall magnitudes are smaller. These differences provide important clues for future experimental tests.

Figure 4. The extrapolated TFFs and versus , where the solid line represents the central value and the shaded bands represent the corresponding uncertainties. In addition, the results are also compared with the predictions from LCSR(I) [75], LCSR(II) [38] and pQCD [36] in the figure.
| 5.726 | 5.726 | 5.324 | 5.279 | |
| 0.338 | 0.188 | 0.518 | 0.560 | |
| 0.460 | 0.519 | |||
| 1.781 | 0.745 | 55.865 | ||
| 0.030% | 0.806% | 0.007% | 0.145% |
Table V. The fitted parameters (GeV), () and quality of extrapolation for TFFs and . All LCSR parameters are set to be their central values.
The differential decay widths are calculated using these TFFs, with adopted [1]. Detailed behavior is shown in Figure 5. To calculate the total decay width of , the integration range over the invariant mass corresponding to Eq. (3) is . The maximum value of given by . Overall, the peak in the differential decay-width distribution for is relatively mild. This behavior can be attributed in part to the Breit-Wigner line shape of the resonance, which redistributes the available phase space and, through the kinematic correlations between and , shifts the distribution toward lower values. In addition, a faster decline is observed in near the endpoint, this is because the phase space of the system is suppressed near the endpoint. To further illustrate the finite-width effects, the double-differential decay width is displayed in the plane in Figure 6. The distribution is mainly concentrated near the resonance region, and the global maximum is found at . As is seen in the figure, when increases from to , the peak position of the distribution is shifted from about to about , indicating that the shape of the distribution is changed with . This behavior arises from the explicit dependence of the kinematic upper limit on : as increases, the kinematically allowed region shrinks. In the finite-width treatment, the resonance is not approximated as a state with a fixed invariant mass; instead, its invariant mass varies across the spectrum. Accounting for the finite width of the resonance is essential for a reliable description of the decay. Our results demonstrate that the finite-width effects of the resonance cannot be neglected in the four-body decay . After integrating over in the whole physical region, the corresponding decay widths can be obtained,

Figure 5. The differential decay widths of are plotted as functions of , the shaded area represents the uncertainty band of for the corresponding channel.

Figure 6. For the process , a three-dimensional distribution of the double differential width as a function of and is presented, where the height of the surface corresponds to the value of the differential decay width.
Finally, using the lifetime and provided by the PDG [1], we have calculated the branching ratio of the decay . To assess the impact of the finite-width effects, we also present the branching fraction for the three-body semileptonic decay in Table VI. For comparison, we include the corresponding theoretical predictions from LCSR (I) [75], LCSR (II) [38], and CLFQM [39]. In Table VI, the branching ratios of in the and channels are almost the same. This indicates that the lepton mass effect effects are negligible, consistent with the expectation of lepton-flavor universality. The branching ratios of the three-body semileptonic decay in the and channels are close to those from LCSR [75]. For the channel, the branching ratio is somewhat smaller than those in the other two channels. This is mainly because the larger mass reduces the available phase space. It is worth noting that, in both the electron and muon channels, the branching fractions for the four-body and three-body semileptonic decays differ by nearly a factor of three. This difference may partly arise from the finite width of the resonance. In the finite-width treatment, the invariant mass squared is integrated over its kinematically allowed range, leading to a four-body decay width that can differ from the result obtained in the narrow-width approximation (NWA) [81]. Currently, no direct experimental measurement of this four-body process is available yet. Our results can provide a reference for future experimental studies.
Table VI. TABLE VI: Predictions for the branching ratios of the decay with and the three-body semileptonic decay are given within the uncertainties (in unit: ).
For multibody decays, a cascade description in terms of sequential decay processes is commonly adopted within the factorization approximation [82, 83, 84]. We include the decay . Under the NWA for the meson, the branching fraction for can then be predicted,
where the [1] and
The above results give the branching fractions for six six-body -meson decay channels arising from the cascade decay involving and . These predictions provide a more complete description of the experimentally accessible final states and extend the phenomenological study of the corresponding multi-body semileptonic decays. They also provide useful theoretical benchmarks for future experimental searches and measurements. Future comparisons between these predictions and experimental data may help test the underlying weak transition dynamics and improve our understanding of the resonance effects in such multi-body decay processes.
Summary
In this paper, we study the decay using the QCD sum rules method. We treat the state as an intermediate resonance in this process and construct its transverse LCDA of the resonance using the LCHO model. The model parameters are then determined by three additional constraints, its behavior is shown in Figure 2. Our model exhibits an asymmetric distribution, which is consistent with the QCDSR [44] predictions. These results can be used as nonperturbative inputs for studying the transition. Using the LCDA determined above, we calculate the TFFs at the maximum-recoil point within the LCSR framework, with the results presented in Table III. We then extrapolate the TFFs over the full physical range to obtain their dependence, as shown in Figure 4. In particular, our prediction for the overall dependence of agrees well with the LCSR(I) predictions [75]. Using the TFFs obtained above, we calculate the differential and total decay widths, as well as the branching fractions, for the decay . The dependence of the differential decay widths is shown in Figure 5. In addition, Figure 6 presents the three-dimensional distribution of the double-differential decay width for as a function of and . The total decay width is obtained by integrating the differential decay width over the full physical range. Meanwhile, using the and lifetimes reported by the PDG [1], we further predict the branching fractions for the four-body decay. To assess the impact of the finite width of the resonance, we also present the branching fractions for the three-body decay within the NWA. The results are summarised in marized in Table VI. As shown in the table, the branching fractions for the four-body decay in the electron and muon channels are approximately 1/3 times of those for the corresponding three-body decay. This indicates that finite-width effects have a substantial impact on the integrated decay rates. Since the can further decay into , the branching ratio of is calculated under the NWA. This provides a theoretical reference for future experiments that observe the decay . Currently, no direct experimental measurement of this process has been reported. It is hoped that in the near future, the results of this work can provide a theoretical reference for future experimental analyses of the final state. They can also provide a theoretical basis for further determining the internal structure of the resonance.
Acknowledgments
This work was supported by the National Natural Science Foundation of China under Grant No.12665017, No.12265010, the Project of Guizhou Provincial Department of Science and Technology under Grants No.MS[2025]219 and No.CXTD[2025]030.
Appendix A
The general analytic expression for the TFFs obtained by the LCSR method is given below:
where , corresponding to , and . Whereas the function with gives the combination . The auxiliary TFF is defined as:
Where is defined as a linear combination of and , it is not an independent physical TFF. To ensure that the hadronic matrix element has no unphysical kinematic singularity at , the condition must be satisfied. Thus, from the relation among , and , we have . The individual coefficients for the two-particle and three-particle distribution amplitudes are given below:
Here , , , and , , with . We further introduce , and . is the usual step function. and are generalized step functions arising from partial integration, their explicit definitions can be found in [60]. The simplified functions and are defined to [85].
References
References
- [1]F. Takahashi et al. [Particle Data Group], Review of Particle Physics, Int. J. Mod. Phys. A 41 (2026) 2630011.
- [2]S. Cheng, A. Khodjamirian and J. Virto, B → ππ Form Factors from Light-Cone Sum Rules with B-meson Distribution Amplitudes, JHEP 05 (2017) 157. [arXiv:1701.01633]
- [3]K. C. Yang, Light-cone distribution amplitudes for the light P-wave singlet state-1 mesons, JHEP 10 (2005) 108. [arXiv:0509337]
- [4]Y. Qiao, Y. X. Liu, Y. G. Xu and R. M. Wang, Study of the axial-vector and tensor resonant contributions to the D → VPℓ+νℓ decays based on SU(3) flavor analysis, Eur. Phys. J. C 84 (2024) 1110. [arXiv:2404.03857]
- [5]M. Ablikim et al. [BESIII Collaboration], Observation of D0 → b1(1235)−e+νe and evidence for D+ → b1(1235)0e+νe, Phys. Rev. Lett. 136 (2026) 021801. [arXiv:2407.20551]
- [6]C. B. Lang, L. Leskovec, D. Mohler and S. Prelovsek, Axial Resonances a1(1260), b1(1235) and Their Decays from the Lattice, JHEP 04 (2014) 162. [arXiv:1401.2088]
- [7]K. Chen, C. Q. Pang, X. Liu and T. Matsuki, Light axial vector mesons, Phys. Rev. D 91 (2015) 074025. [arXiv:1501.07766]
- [8]J. A. Dankowych, P. Brockman, K. W. Edwards, J. Gandsman, D. Legacy, R. S. Longacre, J. F. Martin, P. M. Patel, A. J. Pawlicki and J. D. Prentice, et al. Evidence for I = 1(A1) and I = 0(H) Axial Vector Resonances in Charge Exchange, Phys. Rev. Lett. 46 (1981) 580.
- [9]M. Atkinson et al. [Omega Photon and Bonn-CERN-Glasgow-Lancaster-Manchester-Paris-Rutherford-Sheffield], Photoproduction of π+π−π0 on Hydrogen With Linearly Polarized Photons of Energy 20GeV − 70GeV, Nucl. Phys. B 231 (1984) 15.
- [10]A. Ando, K. Imai, S. Inaba, T. Inagaki, Y. Inagaki, A. Itano, S. Kobayashi, A. Murakami, T. Nakamura and K. Ohmi, et al. Experimental study of the axial vector resonances of a1 and h1 in the π−p charge exchange reaction, Phys. Lett. B 291 (1992) 496.
- [11]D. M. Asner et al. [CLEO Collaboration], Hadronic structure in the decay τ−→π−π−π+π0 and the sign of the Tau Neutrino Helicity, Phys. Rev. D 61 (2000) 012002. [hep-ex/9902022]
- [12]V. Dorofeev, A. Ekimov, Y. Gouz, A. Ivashin, I. Kachaev, A. Karyukhin, Y. Khokhlov, V. Konstantinov, V. Matveev and V. Nikolaenko, et al. Measurement of the f(1)(1285) → π+π−π0 decay, Eur. Phys. J. A 47 (2011) 68.
- [13]M. J. Corden, J. D. Dowell, J. Garvey, M. Jobes, I. R. Kenyon, J. Mawson, T. J. McMahon, I. F. Corbett, R. J. Esterling and N. H. Lipman, et al. Observation of the D0 and Delta Mesons in π−p Interactions at 12GeV/c and 15GeV/c, Nucl. Phys. B 144 (1978) 253.
- [14]A. Gurtu et al. [Amsterdam-CERN-Nijmegen-Oxford], Production and Decay Properties of the D(1285) Meson in K−p Interactions at 4.2GeV/c, Nucl. Phys. B 151 (1979) 181.
- [15]D. Barberis et al. [WA102 Collaboration], A Measurement of the branching fractions of f1(1285) and f1(1420) produced in central pp interactions at 450GeV/c, Phys. Lett. B 440 (1998) 225. [hep-ex/9810003]
- [16]C. Bromberg, J. Dickey, G. Fox, R. Gomez, W. Kropac, J. Pine, S. Stampke, H. Haggerty, E. Malamud and R. Abrams, et al. Observations of the D and E Mesons and Possible Three Kaon Enhancements in π−p → K0K±π∓X, K0K+K−X at 50GeV/c and 100GeV/c, Phys. Rev. D 22 (1980) 1513.
- [17]C. Dionisi et al. [CERN-College de France-Madrid-Stockholm], Observation and Quantum Numbers Determination of the E(1420) Meson in π−p Interactions at 3.95GeV/c, Nucl. Phys. B 169 (1980) 1.
- [18]M. A. Abolins, R. L. Lander, W. A. W. Melhhop, N. huu Xuong and P. M. Yager, Production of Multimeson Resonances via π+p Interaction and Evidence for a πω Resonance, Phys. Rev. Lett. 11 (1963) 381.
- [19]R. Bizzarri, M. Foster, P. Gavillet, G. Labrosse, L. Montanet, R. Salmeron, P. Villemoes, C. Ghesquiere and E. Lillestol, Experimental results on the ωπ and ππ systems as observed in the pp annihilations at rest pp → ω0π+π−, Nucl. Phys. B 14 (1969) 169.
- [20]C. Baltay, J. C. Severiens, N. Yeh and D. Zanello, Observation of the B-meson in the Reaction p̄p → ω0 + π+ + π−, Phys. Rev. Lett. 18 (1967) 93.
- [21]M. Atkinson et al. [Omega Photon Collaboration], Inclusive Photoproduction of Δ(980) and B(1235) at High X(f), Phys. Lett. B 138 (1984) 459.
- [22]P. Weidenauer et al. [ASTERIX Collaboration], N anti-N annihilation at rest into five pions, Z. Phys. C 59 (1993) 387.
- [23]S. Fukui, N. Horikawa, S. Inaba, T. Inagaki, Y. Inagaki, Y. Ishizaki, T. Iwata, T. Kinashi, M. Kurashina and I. Maeda, et al. Study of ω system in the π⁻p charge exchange reaction at 8.95GeV/c, Phys. Lett. B 257 (1991) 241.
- [24]U. Karshon, G. Mikenberg, Y. Eisenberg, S. Pitluck, E. E. Ronat, A. Shapira and G. Yekutieli, Production and Decay Mechanism of the B-meson and the ωπ System in π + p Interactions at 5GeV, Phys. Rev. D 10 (1974) 3608.DOI
- [25]P. Fayet and J. Iliopoulos, Spontaneously Broken Supergauge Symmetries and Goldstone Spinors, Phys. Lett. B 51 (1974) 461.DOI
- [26]R. Gessaroli, A. Quareni-Vignudelli, S. Berti, A. M. Cartacci, C. Caso, F. Fontanelli, S. Squarcia, U. Trevisan, A. Boldetti and G. Costa, et al. Analysis of the ωπ- System Produced in the Reaction π⁻p → π⁻π⁺π⁰p at 11.2GeV/c, Nucl. Phys. B 126 (1977) 382.DOI
- [27]C. Evangelista, B. Ghidini, A. Palano, V. Picciarelli, G. Zito, J. Broring, K. Muller, E. Paul, W. Renneberg and W. Ruhmer, et al. Results From a π⁻p Slow Proton Trigger Experiment at 12 GeV/c, Nucl. Phys. B 178 (1981) 197.DOI
- [28]M. Atkinson et al. [Omega Photon Collaboration], A Spin Parity Analysis of the ωπ⁰ Enhancement Photoproduced in the Energy Range 20GeV to 70GeV, Nucl. Phys. B 243 (1984) 1.DOI
- [29]B. Collick, S. Heppelmann, T. Joyce, Y. Makdisi, M. L. Marshak, E. A. Peterson, K. Ruddick, D. Berg, C. Chandlee and S. Cihangir, et al. PRIMAKOFF PRODUCTION OF THE B+(1235) MESON, Phys. Rev. Lett. 53 (1984) 2374.
- [30]A. Alde et al. [IHEP-IISN-LANL-LAPP-KEK], Study of the omega pi0 system, Z. Phys. C 54 (1992) 553.
- [31]S. Godfrey and N. Isgur, Mesons in a Relativized Quark Model with Chromodynamics, Phys. Rev. D 32 (1985) 189.pubmed.ncbi.nlm.nih.gov/9955999
- [32]D. Ebert, R. N. Faustov and V. O. Galkin, Mass spectra and Regge trajectories of light mesons in the relativistic quark model, Phys. Rev. D 79 (2009) 114029. [arXiv:0903.5183]DOI
- [33]K. Jansen et al. [ETM Collaboration], A Lattice QCD calculation of the transverse decay constant of the b(1)(1235) meson, Phys. Lett. B 690 (2010) 491. [arXiv:0910.5883]DOI
- [34]H. Kim, M. K. Cheoun and K. S. Kim, Spin-1 diquark contributing to the formation of tetraquarks in light mesons, Eur. Phys. J. C 77 (2017) 173. [arXiv:1611.05502]DOI
- [35]P. del Amo Sanchez et al. [BaBar Collaboration], Analysis of the D+ → K−π+e+νe decay channel, Phys. Rev. D 83 (2011) 072001. [arXiv:1201.2180]
- [36]R. H. Li, C. D. Lu and W. Wang, Transition form factors of B decays into P-wave axial-vector mesons in the perturbative QCD approach, Phys. Rev. D 79 (2009) 034014. [arXiv:0901.0307]DOI
- [37]K. C. Yang, Form-Factors of B(u,d,s) Decays into P-Wave Axial-Vector Mesons in the Light-Cone Sum Rule Approach, Phys. Rev. D 78 (2008) 034018. [arXiv:0807.1171]DOI
- [38]Y. J. Sun, Z. G. Wang and T. Huang, B → A Transitions in the Light-Cone QCD Sum Rules with the Chiral current, Chin. Phys. C 36 (2012) 1046.DOI
- [39]X. W. Kang, T. Luo, Y. Zhang, L. Y. Dai and C. Wang, Semileptonic B and B_s decays involving scalar and axial-vector mesons, Eur. Phys. J. C 78 (2018) 909. [arXiv:1808.02432]DOI
- [40]H. Y. Cheng, C. K. Chua and C. W. Hwang, Covariant light front approach for S-wave and P-wave mesons: Its application to decay constants and form-factors, Phys. Rev. D 69 (2004) 074025. [hep-ph/0310359]DOI
- [41]D. Melikhov and B. Stech, Weak form-factors for heavy meson decays: An Update, Phys. Rev. D 62 (2000) 014006. [hep-ph/0001113]DOI
- [42]I. I. Balitsky, V. M. Braun and A. V. Kolesnichenko, Radiative Decay Σ+ → pγ in Quantum Chromodynamics, Nucl. Phys. B 312 (1989) 509-550.DOI
- [43]V. L. Chernyak and I. R. Zhitnitsky, B meson exclusive decays into baryons, Nucl. Phys. B 345 (1990) 137.DOI
- [44]K. C. Yang, Light-cone distribution amplitudes of axial-vector mesons, Nucl. Phys. B 776 (2007) 187-257. [arXiv:0705.0692]DOI
- [45]T. Huang and Z. H. Li, B → K*γ in the light cone QCD sum rule, Phys. Rev. D 57 (1998) 1993.DOI
- [46]T. Huang, Z. H. Li and X. Y. Wu, Improved approach to the heavy to light form-factors in the light cone QCD sum rules, Phys. Rev. D 63 (2001) 094001.DOI
- [47]Z. G. Wang, M. Z. Zhou and T. Huang, B pi weak form-factor with chiral current in the light cone sum rules, Phys. Rev. D 67 (2003) 094006.DOI
- [48]H. B. Fu, X. G. Wu, Y. H. Han and Y. Ma, B → ρ transition form factors and the ρ-meson transverse leading-twist distribution amplitude, J. Phys. G 42 (2015) 055002. [arXiv:1406.3892]
- [49]T. Zhong, X. G. Wu, H. Y. Han, Q. L. Liao, H. B. Fu and Z. Y. Fang, Revisiting the Twist-3 Distribution Amplitudes of K Meson within the QCD Background Field Approach, Commun. Theor. Phys. 58 (2012) 261. [arXiv:1109.3127]DOI
- [50]X. G. Wu, T. Huang and Z. Y. Fang, B → K transition form-factor up to O (1/m_b²) within the k(T) factorization approach, Eur. Phys. J. C 52 (2007) 561. [arXiv:0707.2504]DOI
- [51]T. Huang and X. G. Wu, A Model for the twist-3 wave function of the pion and its contribution to the pion form-factor, Phys. Rev. D 70 (2004) 093013. [arXiv:0408252]DOI
- [52]T. Zhong, H. B. Fu and X. G. Wu, Investigating the ratio of CKM matrix elements |Vub|/|Vcb| from semileptonic decay B_s^0 → K−μ+νμ and kaon twist-2 distribution amplitude, Phys. Rev. D 105 (2022) 116020. [arXiv:2201.10820]
- [53]T. Zhong, Z. H. Zhu, H. B. Fu, X. G. Wu and T. Huang, Improved light-cone harmonic oscillator model for the pionic leading-twist distribution amplitude, Phys. Rev. D 104 (2021) 016021. [arXiv:2102.03989]DOI
- [54]D. D. Hu, X. G. Wu, H. B. Fu, T. Zhong, Z. H. Wu and L. Zeng, Properties of the η_q leading-twist distribution amplitude and its effects to the B/D+ → η(′)ℓ+νℓ decays, Eur. Phys. J. C 84 (2024) 15. [arXiv:2307.04640]DOI
- [55]D. Huang, T. Zhong, H. B. Fu, Z. H. Wu, X. G. Wu and H. Tong, K*0(1430) twist-2 distribution amplitude and B_s, D_s → K*0(1430) transition form factors, Eur. Phys. J. C 83 (2023) 680. [arXiv:2211.06211]DOI
- [56]Y. L. Yang, H. J. Tian, Y. X. Wang, H. B. Fu, T. Zhong, S. Q. Wang and D. Huang, Probing |V_cs| and lepton flavor universality through D → K*0(1430)ℓνℓ decays, Phys. Rev. D 110 (2024) 116030. [arXiv:2409.0512]DOI
- [57]Z. H. Wu, H. B. Fu, T. Zhong, D. Huang, D. D. Hu and X. G. Wu, ω0(980)-meson twist-2 distribution amplitude within the QCD sum rules and investigation of D → a0(980)(ηπ)ℓ+νe, Nucl. Phys. A 1036 (2023) 122671. [arXiv:2211.05390]
- [58]D. D. Hu, X. G. Wu, L. Zeng, H. B. Fu and T. Zhong, Improved light-cone harmonic oscillator model for the ϕ-meson longitudinal leading-twist light-cone distribution amplitude and its effects to D_s^+ → ϕℓ+νℓ, Phys. Rev. D 110 (2024) 056017. [arXiv:2403.10003]
- [59]H. B. Fu, X. G. Wu, W. Cheng and T. Zhong, ρ-meson longitudinal leading-twist distribution amplitude within QCD background field theory, Phys. Rev. D 94 (2016) 074004. [arXiv:1607.04937]DOI
- [60]H. B. Fu, X. G. Wu and Y. Ma, B → K* Transition Form Factors and the Semi-leptonic Decay B → K*μ+μ−, J. Phys. G 43 (2016) 015002. [arXiv:1411.16423]DOI
- [61]D. D. Hu, H. B. Fu, T. Zhong, Z. H. Wu and X. G. Wu, a1(1260)-meson longitudinal twist-2 distribution amplitude and the D → a₁(1260)ℓ⁺νₑ decay processes, Eur. Phys. J. C 82 (2022) 603. [arXiv:2107.02758]
- [62]C. L. Y. Lee, M. Lu and M. B. Wise, B(14) and D(14) decay, Phys. Rev. D 46 (1992) 5040.
- [63]N. Cabibbo and A. Maksymowicz, Angular Correlations in Ke − 4 Decays and Determination of Low-Energy π − π Phase Shifts Phys. Rev. 137 (1965) B438.
- [64]J. D. Richman and P. R. Burchat, Leptonic and semileptonic decays of charm and bottom hadrons, Rev. Mod. Phys. 67 (1995) 893. [hep-ph/9508250]
- [65]M. Ablikim et al. [BESIII Collaboration], Study of D⁺ → K⁻π⁺e⁺νₑ, Phys. Rev. D 94 (2016) 032001. [arXiv:1512.08627]
- [66]P. Ball, V. M. Braun, Y. Koike and K. Tanaka, Higher twist distribution amplitudes of vector mesons in QCD: Formalism and twist - three distributions, Nucl. Phys. B 529 (1998) 323-382. [hep-ph/9802299]DOI
- [67]P. Ball and V. M. Braun, Handbook of higher twist distribution amplitudes of vector mesons in QCD, [hep-ph/9808229].DOI
- [68]P. Ball and G. W. Jones, Twist-3 distribution amplitudes of K* and ϕ mesons, JHEP 03 (2007) 069. [hep-ph/0702100]DOI
- [69]T. Huang, B. Q. Ma and Q. X. Shen, Analysis of the pion wave function in light cone formalism, Phys. Rev. D 49 (1994) 1490-1499. [hep-ph/9402285]DOI
- [70]X. G. Wu and T. Huang, Pion electromagnetic form-factor in the K_T factorization formulae, Int. J. Mod. Phys. A 21 (2006) 901-904. [hep-ph/0507136]DOI
- [71]S. J. Brodsky, T. Huang and G. P. Lepage, The Hadronic Wave Function in Quantum Chromodynamics, SLAC-PUB-2540.
- [72]X. H. Guo and T. Huang, Wave functions in D and B decays, Phys. Rev. D 43 (1991) 2931.
- [73]P. Ball and R. Zwicky, New results on B → π,K,η decay formfactors from light-cone sum rules, Phys. Rev. D 71 (2005) 014015. [hep-ph/0406232]
- [74]V. M. Braun, G. P. Korchemsky and D. Müller, The Uses of conformal symmetry in QCD, Prog. Part. Nucl. Phys. 51 (2003) 311. [hep-ph/0306057]DOI
- [75]L. Zeng, X. G. Wu, D. D. Hu, H. B. Fu and T. Zhong, Longitudinal leading-twist distribution amplitude of the P₁-state b₁(1235) meson and its implications on B → b₁(1235)ℓ⁺νₑ decays, Phys. Rev. D 111 (2025) 056030. [arXiv:2501.06737]DOI
- [76]T. Zhong, Y. H. Dai and H. B. Fu, ρ-meson longitudinal leading-twist distribution amplitude revisited and the D → ρ semileptonic decay, Chin. Phys. C 48 (2024) 063108. [arXiv:2308.14032]arxiv.org/abs/2308.14032
- [77]H. J. Tian, H. B. Fu, T. Zhong, Y. X. Wang and X. G. Wu, Rare decay B⁺ → K⁺ℓ⁺ℓ⁻(ν̄ν) under the QCD sum rules approach, Phys. Rev. D 111 (2025) 076013. [arXiv:2411.12141]arxiv.org/abs/2411.12141
- [78]Y. X. Wang, D. D. Hu, W. B. Luo, T. Zhong and H. B. Fu, Status of the Dₛ⁺ → ϕℓ⁺νₑ decay with a chiral-odd ϕ-meson light-cone distribution amplitude, Phys. Rev. D 112 (2025) 056008. [arXiv:2505.15014]
- [79]C. Bourrely, I. Caprini and L. Lellouch, Model-independent description of B → πℓν decays and determination of |Vᵤb|, Phys. Rev. D 79 (2009) 013008. [arXiv:0807.2722]
- [80]S. Momeni, Helicity form factors for D₍s₎ → Aℓν process in the light-cone QCD sum rules approach, Eur. Phys. J. C 80 (2020) 553. [arXiv:2004.02522]DOI
- [81]H. Y. Cheng, C. W. Chiang and C. K. Chua, Width effects in resonant three-body decays: B decay as an example, Phys. Lett. B 813 (2021) 136058. [arXiv:2011.03201]
- [82]H. Y. Cheng, C. W. Chiang and Z. Q. Zhang, Hadronic three-body D decays mediated by scalar resonances, Phys. Rev. D 105 (2022) 033006. [arXiv:2201.00460]DOI
- [83]J. Wu, N. Wang, G. Lü and X. H. Guo, Study of Kπ final states from four-body decay of B̄ₛ meson under perturbative QCD, Eur. Phys. J. C 85 (2025) 955. [arXiv:2504.18216]
- [84]H. Y. Cheng, C. W. Chiang and C. K. Chua, Finite-Width Effects in Three-Body B Decays, Phys. Rev. D 103 (2021) 036017. [arXiv:2011.07468]DOI
- [85]W. Cheng, X. G. Wu and H. B. Fu, Reconsideration of the B → K* transition form factors within the QCD light-cone sum rules, Phys. Rev. D 95 (2017) 094023. [arXiv:1703.08677]DOI